Ovarian cancer is the eighth most common cancer in women, the fifth most common cause of cancer-related death in women and the leading cause of gynecological cancer death. One- and five-year survival is only 76% and 45%, respectively, primarily due to the late stage at diagnosis for most women. It is clear that risk of invasive epithelial ovarian cancer (ovarian cancer) is driven by both genetic and lifestyle/reproductive factors. Family history is a strong risk factor for ovarian cancer;mutations in BRCA1 and BRCA2 cause ovarian cancer, but account for only ~40% of the excess familial risk of the disease, strongly suggesting that there are other genetic loci to be discovered. A number of risk and protective factors associated with ovarian cancer have been identified, including menopausal hormone therapy use, perineal talc use, obesity, infertility, parity, breast-feeding, oral contraceptive use, and tubal ligation, In this proposal we have the opportunity to explore in a very large dataset the modifying effects of these and other lifestyle and reproductive factors on ovarian cancer genetic susceptibility loci that have been identified through our recently completed genome-wide association study (GWAS). In stage 1 of our GWAS we have studied 550,000 single nucleotide polymorphisms (SNPs) in 2000 ovarian cancer cases and ~1500 controls. The top 28,219-associated SNPs have been studied in an additional 5,145 cases and 5,506 controls and we have also compiled a common dataset of epidemiological variables for all of these samples. We will examine this dataset for evidence of gene- lifestyle/reproductive factor interactions. In addition, we will fine map the five regions that have now definitively been shown to harbor an ovarian cancer susceptibility allele in order to identify the set of possible causal variants. This work is possible through the collaborative efforts of more than 20 leading research groups from around the world who are committed to improving risk prediction and early stage disease detection. Women diagnosed with ovarian cancer at an early stage, which currently represents only 30% of cases, have a five year survival of 90%. Identifying novel causal variants and the manner in which known ovarian cancer risk and protective factors interact with genetic variants would lead to a major improvement in our understanding of the disease and ultimately improvements in prevention and survival. PUBLIC HEALTH RELEVANCE: Ovarian cancer is the eighth most common cancer in women, the fifth most common cause of cancer-related death in women and the leading cause of gynecological cancer death;one- and five-year survival is a paltry 76% and 45%, respectively. We will utilize the data generated through our genome-wide association study (GWAS) in 7,145 cases and 5,506 controls to identify gene-environment interactions for 13 important ovarian cancer risk and protective factors. We will also fine map the five regions definitively shown to harbor an ovarian cancer susceptibility alleles and address a key methodological challenge faced in studies of these kinds.